Markov Model-Based Reliability Analysis Considering the Redundancy Effect of Modular Converters

Abstract

This paper uses a DAB converter and a non-isolated DC-DC converter as modules. It analyzes the differences in circuit characteristics and reliability due to the redundancy effect by combining them in series and parallel. In particular, reliability is analyzed considering the redundancy effect when the rated power of the DAB converter and non-isolated DC-DC converter are designed to be 50% and 100% of the module-rated power, respectively. The conventional FTA method can analyze failures that reflect the operational risks of modules but cannot analyze partial failures when failures occur in some modules. To solve this problem, this paper predicts the failure rate and MTBF of the converter by modeling state changes and partial failures between modules of a modular converter based on the Markov model. The impact of the redundancy effect on reliability is presented as a quantitative value by calculating the failure rate of major components from the MIL-STD-217F fault library and substituting it into the Markov model.

Keywords

• Bidirectional DC–DC converters
• dual active bridge (DAB)
• fault tree analysis (FTA)
• Markov model
• mean time between failures (MTBF)
• redundancy